[deliverable/linux.git] / arch / x86 / xen / setup.c

/*
 * Machine specific setup for xen
 *
 * Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007
 */

#include <linux/module.h>
#include <linux/sched.h>
#include <linux/mm.h>
#include <linux/pm.h>
#include <linux/memblock.h>

#include <asm/elf.h>
#include <asm/vdso.h>
#include <asm/e820.h>
#include <asm/setup.h>
#include <asm/acpi.h>
#include <asm/xen/hypervisor.h>
#include <asm/xen/hypercall.h>

#include <xen/xen.h>
#include <xen/page.h>
#include <xen/interface/callback.h>
#include <xen/interface/memory.h>
#include <xen/interface/physdev.h>
#include <xen/features.h>

#include "xen-ops.h"
#include "vdso.h"

/* These are code, but not functions.  Defined in entry.S */
extern const char xen_hypervisor_callback[];
extern const char xen_failsafe_callback[];
extern void xen_sysenter_target(void);
extern void xen_syscall_target(void);
extern void xen_syscall32_target(void);

/* Amount of extra memory space we add to the e820 ranges */
phys_addr_t xen_extra_mem_start, xen_extra_mem_size;

/* 
 * The maximum amount of extra memory compared to the base size.  The
 * main scaling factor is the size of struct page.  At extreme ratios
 * of base:extra, all the base memory can be filled with page
 * structures for the extra memory, leaving no space for anything
 * else.
 * 
 * 10x seems like a reasonable balance between scaling flexibility and
 * leaving a practically usable system.
 */
#define EXTRA_MEM_RATIO		(10)

static __init void xen_add_extra_mem(unsigned long pages)
{
	u64 size = (u64)pages * PAGE_SIZE;
	u64 extra_start = xen_extra_mem_start + xen_extra_mem_size;

	if (!pages)
		return;

	e820_add_region(extra_start, size, E820_RAM);
	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);

	memblock_x86_reserve_range(extra_start, extra_start + size, "XEN EXTRA");

	xen_extra_mem_size += size;

	xen_max_p2m_pfn = PFN_DOWN(extra_start + size);
}

static unsigned long __init xen_release_chunk(phys_addr_t start_addr,
					      phys_addr_t end_addr)
{
	struct xen_memory_reservation reservation = {
		.address_bits = 0,
		.extent_order = 0,
		.domid        = DOMID_SELF
	};
	unsigned long start, end;
	unsigned long len = 0;
	unsigned long pfn;
	int ret;

	start = PFN_UP(start_addr);
	end = PFN_DOWN(end_addr);

	if (end <= start)
		return 0;

	printk(KERN_INFO "xen_release_chunk: looking at area pfn %lx-%lx: ",
	       start, end);
	for(pfn = start; pfn < end; pfn++) {
		unsigned long mfn = pfn_to_mfn(pfn);

		/* Make sure pfn exists to start with */
		if (mfn == INVALID_P2M_ENTRY || mfn_to_pfn(mfn) != pfn)
			continue;

		set_xen_guest_handle(reservation.extent_start, &mfn);
		reservation.nr_extents = 1;

		ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
					   &reservation);
		WARN(ret != 1, "Failed to release memory %lx-%lx err=%d\n",
		     start, end, ret);
		if (ret == 1) {
			set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
			len++;
		}
	}
	printk(KERN_CONT "%ld pages freed\n", len);

	return len;
}

static unsigned long __init xen_return_unused_memory(unsigned long max_pfn,
						     const struct e820map *e820)
{
	phys_addr_t max_addr = PFN_PHYS(max_pfn);
	phys_addr_t last_end = ISA_END_ADDRESS;
	unsigned long released = 0;
	int i;

	/* Free any unused memory above the low 1Mbyte. */
	for (i = 0; i < e820->nr_map && last_end < max_addr; i++) {
		phys_addr_t end = e820->map[i].addr;
		end = min(max_addr, end);

		if (last_end < end)
			released += xen_release_chunk(last_end, end);
		last_end = max(last_end, e820->map[i].addr + e820->map[i].size);
	}

	if (last_end < max_addr)
		released += xen_release_chunk(last_end, max_addr);

	printk(KERN_INFO "released %ld pages of unused memory\n", released);
	return released;
}

/**
 * machine_specific_memory_setup - Hook for machine specific memory setup.
 **/
char * __init xen_memory_setup(void)
{
	static struct e820entry map[E820MAX] __initdata;

	unsigned long max_pfn = xen_start_info->nr_pages;
	unsigned long long mem_end;
	int rc;
	struct xen_memory_map memmap;
	unsigned long extra_pages = 0;
	unsigned long extra_limit;
	int i;
	int op;

	max_pfn = min(MAX_DOMAIN_PAGES, max_pfn);
	mem_end = PFN_PHYS(max_pfn);

	memmap.nr_entries = E820MAX;
	set_xen_guest_handle(memmap.buffer, map);

	op = xen_initial_domain() ?
		XENMEM_machine_memory_map :
		XENMEM_memory_map;
	rc = HYPERVISOR_memory_op(op, &memmap);
	if (rc == -ENOSYS) {
		BUG_ON(xen_initial_domain());
		memmap.nr_entries = 1;
		map[0].addr = 0ULL;
		map[0].size = mem_end;
		/* 8MB slack (to balance backend allocations). */
		map[0].size += 8ULL << 20;
		map[0].type = E820_RAM;
		rc = 0;
	}
	BUG_ON(rc);

	e820.nr_map = 0;
	xen_extra_mem_start = mem_end;
	for (i = 0; i < memmap.nr_entries; i++) {
		unsigned long long end;

		/* Guard against non-page aligned E820 entries. */
		if (map[i].type == E820_RAM)
			map[i].size -= (map[i].size + map[i].addr) % PAGE_SIZE;

		end = map[i].addr + map[i].size;
		if (map[i].type == E820_RAM && end > mem_end) {
			/* RAM off the end - may be partially included */
			u64 delta = min(map[i].size, end - mem_end);

			map[i].size -= delta;
			end -= delta;

			extra_pages += PFN_DOWN(delta);
		}

		if (map[i].size > 0 && end > xen_extra_mem_start)
			xen_extra_mem_start = end;

		/* Add region if any remains */
		if (map[i].size > 0)
			e820_add_region(map[i].addr, map[i].size, map[i].type);
	}

	/*
	 * In domU, the ISA region is normal, usable memory, but we
	 * reserve ISA memory anyway because too many things poke
	 * about in there.
	 *
	 * In Dom0, the host E820 information can leave gaps in the
	 * ISA range, which would cause us to release those pages.  To
	 * avoid this, we unconditionally reserve them here.
	 */
	e820_add_region(ISA_START_ADDRESS, ISA_END_ADDRESS - ISA_START_ADDRESS,
			E820_RESERVED);

	/*
	 * Reserve Xen bits:
	 *  - mfn_list
	 *  - xen_start_info
	 * See comment above "struct start_info" in <xen/interface/xen.h>
	 */
	memblock_x86_reserve_range(__pa(xen_start_info->mfn_list),
		      __pa(xen_start_info->pt_base),
			"XEN START INFO");

	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);

	extra_pages += xen_return_unused_memory(xen_start_info->nr_pages, &e820);

	/*
	 * Clamp the amount of extra memory to a EXTRA_MEM_RATIO
	 * factor the base size.  On non-highmem systems, the base
	 * size is the full initial memory allocation; on highmem it
	 * is limited to the max size of lowmem, so that it doesn't
	 * get completely filled.
	 *
	 * In principle there could be a problem in lowmem systems if
	 * the initial memory is also very large with respect to
	 * lowmem, but we won't try to deal with that here.
	 */
	extra_limit = min(EXTRA_MEM_RATIO * min(max_pfn, PFN_DOWN(MAXMEM)),
			  max_pfn + extra_pages);

	if (extra_limit >= max_pfn)
		extra_pages = extra_limit - max_pfn;
	else
		extra_pages = 0;

	xen_add_extra_mem(extra_pages);

	return "Xen";
}

/*
 * Set the bit indicating "nosegneg" library variants should be used.
 * We only need to bother in pure 32-bit mode; compat 32-bit processes
 * can have un-truncated segments, so wrapping around is allowed.
 */
static void __init fiddle_vdso(void)
{
#ifdef CONFIG_X86_32
	u32 *mask;
	mask = VDSO32_SYMBOL(&vdso32_int80_start, NOTE_MASK);
	*mask |= 1 << VDSO_NOTE_NONEGSEG_BIT;
	mask = VDSO32_SYMBOL(&vdso32_sysenter_start, NOTE_MASK);
	*mask |= 1 << VDSO_NOTE_NONEGSEG_BIT;
#endif
}

static __cpuinit int register_callback(unsigned type, const void *func)
{
	struct callback_register callback = {
		.type = type,
		.address = XEN_CALLBACK(__KERNEL_CS, func),
		.flags = CALLBACKF_mask_events,
	};

	return HYPERVISOR_callback_op(CALLBACKOP_register, &callback);
}

void __cpuinit xen_enable_sysenter(void)
{
	int ret;
	unsigned sysenter_feature;

#ifdef CONFIG_X86_32
	sysenter_feature = X86_FEATURE_SEP;
#else
	sysenter_feature = X86_FEATURE_SYSENTER32;
#endif

	if (!boot_cpu_has(sysenter_feature))
		return;

	ret = register_callback(CALLBACKTYPE_sysenter, xen_sysenter_target);
	if(ret != 0)
		setup_clear_cpu_cap(sysenter_feature);
}

void __cpuinit xen_enable_syscall(void)
{
#ifdef CONFIG_X86_64
	int ret;

	ret = register_callback(CALLBACKTYPE_syscall, xen_syscall_target);
	if (ret != 0) {
		printk(KERN_ERR "Failed to set syscall callback: %d\n", ret);
		/* Pretty fatal; 64-bit userspace has no other
		   mechanism for syscalls. */
	}

	if (boot_cpu_has(X86_FEATURE_SYSCALL32)) {
		ret = register_callback(CALLBACKTYPE_syscall32,
					xen_syscall32_target);
		if (ret != 0)
			setup_clear_cpu_cap(X86_FEATURE_SYSCALL32);
	}
#endif /* CONFIG_X86_64 */
}

void __init xen_arch_setup(void)
{
	xen_panic_handler_init();

	HYPERVISOR_vm_assist(VMASST_CMD_enable, VMASST_TYPE_4gb_segments);
	HYPERVISOR_vm_assist(VMASST_CMD_enable, VMASST_TYPE_writable_pagetables);

	if (!xen_feature(XENFEAT_auto_translated_physmap))
		HYPERVISOR_vm_assist(VMASST_CMD_enable,
				     VMASST_TYPE_pae_extended_cr3);

	if (register_callback(CALLBACKTYPE_event, xen_hypervisor_callback) ||
	    register_callback(CALLBACKTYPE_failsafe, xen_failsafe_callback))
		BUG();

	xen_enable_sysenter();
	xen_enable_syscall();

#ifdef CONFIG_ACPI
	if (!(xen_start_info->flags & SIF_INITDOMAIN)) {
		printk(KERN_INFO "ACPI in unprivileged domain disabled\n");
		disable_acpi();
	}
#endif

	memcpy(boot_command_line, xen_start_info->cmd_line,
	       MAX_GUEST_CMDLINE > COMMAND_LINE_SIZE ?
	       COMMAND_LINE_SIZE : MAX_GUEST_CMDLINE);

	/* Set up idle, making sure it calls safe_halt() pvop */
#ifdef CONFIG_X86_32
	boot_cpu_data.hlt_works_ok = 1;
#endif
	pm_idle = default_idle;
	boot_option_idle_override = IDLE_HALT;

	fiddle_vdso();
}
Commit	Line	Data
5ead97c8 JF	1	/*
	2	* Machine specific setup for xen
	3	*
	4	* Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007
	5	*/
	6
	7	#include <linux/module.h>
	8	#include <linux/sched.h>
	9	#include <linux/mm.h>
	10	#include <linux/pm.h>
a9ce6bc1	11	#include <linux/memblock.h>
5ead97c8 JF	12
5ead97c8 JF	13	#include <asm/elf.h>
6c3652ef	14	#include <asm/vdso.h>
5ead97c8 JF	15	#include <asm/e820.h>
5ead97c8 JF	16	#include <asm/setup.h>
b792c755	17	#include <asm/acpi.h>
5ead97c8 JF	18	#include <asm/xen/hypervisor.h>
	19	#include <asm/xen/hypercall.h>
	20
45263cb0	21	#include <xen/xen.h>
8006ec3e	22	#include <xen/page.h>
e2a81baf	23	#include <xen/interface/callback.h>
35ae11fd	24	#include <xen/interface/memory.h>
5ead97c8 JF	25	#include <xen/interface/physdev.h>
	26	#include <xen/features.h>
	27
	28	#include "xen-ops.h"
d2eea68e	29	#include "vdso.h"
5ead97c8 JF	30
	31	/* These are code, but not functions. Defined in entry.S */
	32	extern const char xen_hypervisor_callback[];
	33	extern const char xen_failsafe_callback[];
f63c2f24 T	34	extern void xen_sysenter_target(void);
	35	extern void xen_syscall_target(void);
	36	extern void xen_syscall32_target(void);
5ead97c8	37
42ee1471 JF	38	/* Amount of extra memory space we add to the e820 ranges */
	39	phys_addr_t xen_extra_mem_start, xen_extra_mem_size;
	40
698bb8d1 JF	41	/*
	42	* The maximum amount of extra memory compared to the base size. The
	43	* main scaling factor is the size of struct page. At extreme ratios
	44	* of base:extra, all the base memory can be filled with page
	45	* structures for the extra memory, leaving no space for anything
	46	* else.
	47	*
	48	* 10x seems like a reasonable balance between scaling flexibility and
	49	* leaving a practically usable system.
	50	*/
	51	#define EXTRA_MEM_RATIO (10)
	52
42ee1471 JF	53	static __init void xen_add_extra_mem(unsigned long pages)
	54	{
	55	u64 size = (u64)pages * PAGE_SIZE;
3654581e	56	u64 extra_start = xen_extra_mem_start + xen_extra_mem_size;
42ee1471 JF	57
	58	if (!pages)
	59	return;
	60
3654581e	61	e820_add_region(extra_start, size, E820_RAM);
42ee1471 JF	62	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
42ee1471 JF	63
520045db	64	memblock_x86_reserve_range(extra_start, extra_start + size, "XEN EXTRA");
42ee1471 JF	65
42ee1471 JF	66	xen_extra_mem_size += size;
2f7acb20	67
3654581e	68	xen_max_p2m_pfn = PFN_DOWN(extra_start + size);
42ee1471 JF	69	}
42ee1471 JF	70
f89e048e JF	71	static unsigned long __init xen_release_chunk(phys_addr_t start_addr,
f89e048e JF	72	phys_addr_t end_addr)
093d7b46 MR	73	{
	74	struct xen_memory_reservation reservation = {
	75	.address_bits = 0,
	76	.extent_order = 0,
	77	.domid = DOMID_SELF
	78	};
093d7b46	79	unsigned long start, end;
f89e048e	80	unsigned long len = 0;
093d7b46 MR	81	unsigned long pfn;
	82	int ret;
	83
	84	start = PFN_UP(start_addr);
f89e048e	85	end = PFN_DOWN(end_addr);
093d7b46 MR	86
	87	if (end <= start)
	88	return 0;
	89
f89e048e JF	90	printk(KERN_INFO "xen_release_chunk: looking at area pfn %lx-%lx: ",
	91	start, end);
	92	for(pfn = start; pfn < end; pfn++) {
	93	unsigned long mfn = pfn_to_mfn(pfn);
	94
	95	/* Make sure pfn exists to start with */
	96	if (mfn == INVALID_P2M_ENTRY \|\| mfn_to_pfn(mfn) != pfn)
	97	continue;
	98
	99	set_xen_guest_handle(reservation.extent_start, &mfn);
	100	reservation.nr_extents = 1;
	101
	102	ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
	103	&reservation);
	104	WARN(ret != 1, "Failed to release memory %lx-%lx err=%d\n",
	105	start, end, ret);
	106	if (ret == 1) {
	107	set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
	108	len++;
	109	}
	110	}
	111	printk(KERN_CONT "%ld pages freed\n", len);
093d7b46 MR	112
	113	return len;
	114	}
	115
f89e048e JF	116	static unsigned long __init xen_return_unused_memory(unsigned long max_pfn,
f89e048e JF	117	const struct e820map *e820)
093d7b46	118	{
f89e048e	119	phys_addr_t max_addr = PFN_PHYS(max_pfn);
9ec23a7f	120	phys_addr_t last_end = ISA_END_ADDRESS;
093d7b46 MR	121	unsigned long released = 0;
	122	int i;
	123
9ec23a7f	124	/* Free any unused memory above the low 1Mbyte. */
f89e048e JF	125	for (i = 0; i < e820->nr_map && last_end < max_addr; i++) {
	126	phys_addr_t end = e820->map[i].addr;
	127	end = min(max_addr, end);
	128
9ec23a7f IC	129	if (last_end < end)
	130	released += xen_release_chunk(last_end, end);
	131	last_end = max(last_end, e820->map[i].addr + e820->map[i].size);
093d7b46 MR	132	}
093d7b46 MR	133
f89e048e JF	134	if (last_end < max_addr)
f89e048e JF	135	released += xen_release_chunk(last_end, max_addr);
093d7b46 MR	136
	137	printk(KERN_INFO "released %ld pages of unused memory\n", released);
	138	return released;
	139	}
5ead97c8 JF	140
	141	/**
	142	* machine_specific_memory_setup - Hook for machine specific memory setup.
	143	**/
5ead97c8 JF	144	char * __init xen_memory_setup(void)
5ead97c8 JF	145	{
35ae11fd IC	146	static struct e820entry map[E820MAX] __initdata;
35ae11fd IC	147
5ead97c8	148	unsigned long max_pfn = xen_start_info->nr_pages;
35ae11fd IC	149	unsigned long long mem_end;
	150	int rc;
	151	struct xen_memory_map memmap;
42ee1471	152	unsigned long extra_pages = 0;
698bb8d1	153	unsigned long extra_limit;
35ae11fd	154	int i;
9e9a5fcb	155	int op;
5ead97c8	156
8006ec3e	157	max_pfn = min(MAX_DOMAIN_PAGES, max_pfn);
35ae11fd IC	158	mem_end = PFN_PHYS(max_pfn);
	159
	160	memmap.nr_entries = E820MAX;
	161	set_xen_guest_handle(memmap.buffer, map);
	162
9e9a5fcb IC	163	op = xen_initial_domain() ?
	164	XENMEM_machine_memory_map :
	165	XENMEM_memory_map;
	166	rc = HYPERVISOR_memory_op(op, &memmap);
35ae11fd	167	if (rc == -ENOSYS) {
9ec23a7f	168	BUG_ON(xen_initial_domain());
35ae11fd IC	169	memmap.nr_entries = 1;
	170	map[0].addr = 0ULL;
	171	map[0].size = mem_end;
	172	/* 8MB slack (to balance backend allocations). */
	173	map[0].size += 8ULL << 20;
	174	map[0].type = E820_RAM;
	175	rc = 0;
	176	}
	177	BUG_ON(rc);
8006ec3e	178
5ead97c8	179	e820.nr_map = 0;
42ee1471	180	xen_extra_mem_start = mem_end;
35ae11fd	181	for (i = 0; i < memmap.nr_entries; i++) {
7cb31b75	182	unsigned long long end;
be5bf9fa	183
7cb31b75 SS	184	/* Guard against non-page aligned E820 entries. */
	185	if (map[i].type == E820_RAM)
	186	map[i].size -= (map[i].size + map[i].addr) % PAGE_SIZE;
	187
	188	end = map[i].addr + map[i].size;
c2d08791 JF	189	if (map[i].type == E820_RAM && end > mem_end) {
	190	/* RAM off the end - may be partially included */
	191	u64 delta = min(map[i].size, end - mem_end);
42ee1471	192
c2d08791 JF	193	map[i].size -= delta;
c2d08791 JF	194	end -= delta;
3654581e	195
c2d08791	196	extra_pages += PFN_DOWN(delta);
3654581e JF	197	}
3654581e JF	198
c2d08791	199	if (map[i].size > 0 && end > xen_extra_mem_start)
36bc251b	200	xen_extra_mem_start = end;
b5b43ced	201
c2d08791 JF	202	/* Add region if any remains */
c2d08791 JF	203	if (map[i].size > 0)
35ae11fd IC	204	e820_add_region(map[i].addr, map[i].size, map[i].type);
35ae11fd IC	205	}
b792c755 JF	206
b792c755 JF	207	/*
9ec23a7f IC	208	* In domU, the ISA region is normal, usable memory, but we
9ec23a7f IC	209	* reserve ISA memory anyway because too many things poke
b792c755	210	* about in there.
4ec5387c	211	*
9ec23a7f IC	212	* In Dom0, the host E820 information can leave gaps in the
	213	* ISA range, which would cause us to release those pages. To
	214	* avoid this, we unconditionally reserve them here.
b792c755 JF	215	*/
	216	e820_add_region(ISA_START_ADDRESS, ISA_END_ADDRESS - ISA_START_ADDRESS,
	217	E820_RESERVED);
5ead97c8	218
be5bf9fa JF	219	/*
	220	* Reserve Xen bits:
	221	* - mfn_list
	222	* - xen_start_info
	223	* See comment above "struct start_info" in <xen/interface/xen.h>
	224	*/
a9ce6bc1	225	memblock_x86_reserve_range(__pa(xen_start_info->mfn_list),
6b2e8523 JF	226	__pa(xen_start_info->pt_base),
6b2e8523 JF	227	"XEN START INFO");
be5bf9fa JF	228
	229	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
	230
42ee1471 JF	231	extra_pages += xen_return_unused_memory(xen_start_info->nr_pages, &e820);
42ee1471 JF	232
698bb8d1 JF	233	/*
	234	* Clamp the amount of extra memory to a EXTRA_MEM_RATIO
	235	* factor the base size. On non-highmem systems, the base
	236	* size is the full initial memory allocation; on highmem it
	237	* is limited to the max size of lowmem, so that it doesn't
	238	* get completely filled.
	239	*
	240	* In principle there could be a problem in lowmem systems if
	241	* the initial memory is also very large with respect to
	242	* lowmem, but we won't try to deal with that here.
	243	*/
	244	extra_limit = min(EXTRA_MEM_RATIO * min(max_pfn, PFN_DOWN(MAXMEM)),
	245	max_pfn + extra_pages);
	246
	247	if (extra_limit >= max_pfn)
	248	extra_pages = extra_limit - max_pfn;
	249	else
	250	extra_pages = 0;
	251
d2a81713	252	xen_add_extra_mem(extra_pages);
093d7b46	253
5ead97c8 JF	254	return "Xen";
	255	}
	256
d2eea68e RM	257	/*
d2eea68e RM	258	* Set the bit indicating "nosegneg" library variants should be used.
6a52e4b1 JF	259	* We only need to bother in pure 32-bit mode; compat 32-bit processes
6a52e4b1 JF	260	* can have un-truncated segments, so wrapping around is allowed.
d2eea68e	261	*/
08b6d290	262	static void __init fiddle_vdso(void)
d2eea68e	263	{
6a52e4b1 JF	264	#ifdef CONFIG_X86_32
	265	u32 *mask;
	266	mask = VDSO32_SYMBOL(&vdso32_int80_start, NOTE_MASK);
	267	*mask \|= 1 << VDSO_NOTE_NONEGSEG_BIT;
	268	mask = VDSO32_SYMBOL(&vdso32_sysenter_start, NOTE_MASK);
d2eea68e	269	*mask \|= 1 << VDSO_NOTE_NONEGSEG_BIT;
6fcac6d3	270	#endif
d2eea68e RM	271	}
d2eea68e RM	272
88459d4c	273	static __cpuinit int register_callback(unsigned type, const void *func)
e2a81baf	274	{
88459d4c JF	275	struct callback_register callback = {
	276	.type = type,
	277	.address = XEN_CALLBACK(__KERNEL_CS, func),
e2a81baf JF	278	.flags = CALLBACKF_mask_events,
	279	};
	280
88459d4c JF	281	return HYPERVISOR_callback_op(CALLBACKOP_register, &callback);
	282	}
	283
	284	void __cpuinit xen_enable_sysenter(void)
	285	{
6fcac6d3	286	int ret;
62541c37	287	unsigned sysenter_feature;
6fcac6d3 JF	288
6fcac6d3 JF	289	#ifdef CONFIG_X86_32
62541c37	290	sysenter_feature = X86_FEATURE_SEP;
6fcac6d3	291	#else
62541c37	292	sysenter_feature = X86_FEATURE_SYSENTER32;
6fcac6d3	293	#endif
88459d4c	294
62541c37 JF	295	if (!boot_cpu_has(sysenter_feature))
	296	return;
	297
6fcac6d3	298	ret = register_callback(CALLBACKTYPE_sysenter, xen_sysenter_target);
62541c37 JF	299	if(ret != 0)
62541c37 JF	300	setup_clear_cpu_cap(sysenter_feature);
e2a81baf JF	301	}
e2a81baf JF	302
6fcac6d3 JF	303	void __cpuinit xen_enable_syscall(void)
	304	{
	305	#ifdef CONFIG_X86_64
6fcac6d3	306	int ret;
6fcac6d3 JF	307
	308	ret = register_callback(CALLBACKTYPE_syscall, xen_syscall_target);
	309	if (ret != 0) {
d5303b81	310	printk(KERN_ERR "Failed to set syscall callback: %d\n", ret);
62541c37 JF	311	/* Pretty fatal; 64-bit userspace has no other
	312	mechanism for syscalls. */
	313	}
	314
	315	if (boot_cpu_has(X86_FEATURE_SYSCALL32)) {
6fcac6d3 JF	316	ret = register_callback(CALLBACKTYPE_syscall32,
6fcac6d3 JF	317	xen_syscall32_target);
d5303b81	318	if (ret != 0)
62541c37	319	setup_clear_cpu_cap(X86_FEATURE_SYSCALL32);
6fcac6d3 JF	320	}
	321	#endif /* CONFIG_X86_64 */
	322	}
	323
5ead97c8 JF	324	void __init xen_arch_setup(void)
5ead97c8 JF	325	{
f09f6d19 DD	326	xen_panic_handler_init();
f09f6d19 DD	327
5ead97c8 JF	328	HYPERVISOR_vm_assist(VMASST_CMD_enable, VMASST_TYPE_4gb_segments);
	329	HYPERVISOR_vm_assist(VMASST_CMD_enable, VMASST_TYPE_writable_pagetables);
	330
	331	if (!xen_feature(XENFEAT_auto_translated_physmap))
f63c2f24 T	332	HYPERVISOR_vm_assist(VMASST_CMD_enable,
f63c2f24 T	333	VMASST_TYPE_pae_extended_cr3);
5ead97c8	334
88459d4c JF	335	if (register_callback(CALLBACKTYPE_event, xen_hypervisor_callback) \|\|
	336	register_callback(CALLBACKTYPE_failsafe, xen_failsafe_callback))
	337	BUG();
5ead97c8	338
e2a81baf	339	xen_enable_sysenter();
6fcac6d3	340	xen_enable_syscall();
e2a81baf	341
5ead97c8 JF	342	#ifdef CONFIG_ACPI
	343	if (!(xen_start_info->flags & SIF_INITDOMAIN)) {
	344	printk(KERN_INFO "ACPI in unprivileged domain disabled\n");
	345	disable_acpi();
	346	}
	347	#endif
	348
	349	memcpy(boot_command_line, xen_start_info->cmd_line,
	350	MAX_GUEST_CMDLINE > COMMAND_LINE_SIZE ?
	351	COMMAND_LINE_SIZE : MAX_GUEST_CMDLINE);
	352
bc15fde7 JF	353	/* Set up idle, making sure it calls safe_halt() pvop */
	354	#ifdef CONFIG_X86_32
	355	boot_cpu_data.hlt_works_ok = 1;
	356	#endif
	357	pm_idle = default_idle;
23febedd	358	boot_option_idle_override = IDLE_HALT;
f87e4cac	359
d2eea68e	360	fiddle_vdso();
5ead97c8	361	}